E/n/r/i.\io, 3X(I I ) : I 157-1 166, 1997 Lippincolt-Ravcn Publishers, Philadelphia 0 Inlernntional Lcaguc Against Epilepsy

Multiple Kainic Acid Seizures in the Immature and Adult Brain: Ictal Manifestations and Long-Term Effects on Learning

and Memory

*Matthew R. Sarkisian, "Pushpa Tandon, *Zhao Liu, "Yili Yang, "Ariyuki Hori, *Gregory L. Holmes, and *?Carl E. Stafstrom

*Department of Neurology, Harvard Medical School, Children's Hospital, and fDivision of Pediatric Neurology, Tufts Universiry School of Medicine, New England Medical Center, Boston, Massachusetts, U.S.A.

Summary: Purpose: While there is increasing evidence that the adverse effects of prolonged seizures are less pronounced i n the immature than in the mature brain, there have been few investigations of the long-term effects of recurrent seizures during development. This study examined the effects of mul- tiple administrations of the convulsant kainic acid (KA) on seizure characteristics and spatial learning as a function of brain development.

Metho& To determine the long-term effects of serial KA seizures during ontogeny, saline or convulsant doses of KA were given intraperitoneally 4 times, at 2-day intervals. Imma- ture rats were given KA on P20, P22, P24 and P26; adult rats got KA on P60, P62, P64 and P66. Ictal characteristics and EEGs were recorded. To examine the effects of multiple KA seizures on the retention of spatial learning, water maze testing was performed before (immature group: from P16-19, adult group: from P56-P59) and after (immature: from P60-P63, adult: from PIOO-Pl03) KA injections. Finally, histology was performed to compare KA-induced damage at each age.

Results: In immature animals, serial KA administration re- sulted in seizures with a progressively longer onset latency and

decreased severity. In contrast, KA serially administered to adult rats caused severe seizures after each of the 4 injections. In immature rats, epileptiform EEG changes were most promi- nent after the first KA injection, whereas in adults, prolonged paroxysmal EEG patterns were seen after all 4 KA injections. Before KA, both rat pups and adults acquired place learning in the water maze. One month after the final KA injection, there was no deficit in spatial learning retention in the immature group, whereas the adult group had profound impairment com- pared to age-matched, saline-injected controls. Histology re- vealed no lesions in immature rats treated multiple times with KA but profound cell loss in hippocampal fields CA4, CA3 and CAI in rats treated serially with KA as adults.

Conclusions: Previous studies have shown that a single KA injection causes prolonged status epilepticus (which persists for several hours), leading to severe histologic and behavioral se- quelae in adult rats but not in pups. Our study extends those findings, demonstrating that immature rats are spared the cog- nitive and pathological sequelae of multiple injections of con- vulsant doses of KA as well. Key Words: Seizures-Kainic acid-Immature brain-Water maze-Learning.

Research using a variety of animal models has dem- onstrated that the long-term effects of seizures are age- dependent. Kainic acid (KA), an analogue of the natu- rally occurring excitatory amino acid neurotransmitter glutamate, when given either systemically or intracere- brally to adult rats, results in a condition similar to hu- man temporal lobe epilepsy (1,2). KA-induced status epilepticus causes long-term deficits in learning, memory, behavior and increased seizure susceptibility in mature rats, but has no discernible effects in rats under- going KA status epilepticus at ages < about 20 days (3-6). In addition, in adult rats, KA seizures cause severe

Accepted May 13, 1997 Address correspondence and reprint requests to Dr. C. E. Stafstrom

at Division of Pediatric Neurology, #330, New England Medical Cen- ter, 750 Washington Street, Boston, MA 02111, U.S.A.

neuronal loss in hippocampal fields CA1, CA3, dentate hilus and synaptic reorganization in the form of mossy fiber sprouting; no such changes are noted in immature rat pups after KA status epilepticus (5). In other experi- mental epilepsy models, the immature brain is similarly spared seizure-induced brain damage. Continuous hippo- campal stimulation, a model of limbic status epilepticus, causes long-term adverse effects on cognition in pubes- cent and mature rats but not in prepubescent animals (7). Similarly, pilocarpine administration, which produces severe seizures in both immature and mature animals, results in fewer behavioral abnormalities and less histo- logic damage in immature animals than in adults (8). Thus, the age at which status epilepticus occurs is a crucial determinant of whether or not long-term impair- ment in learning and memory will occur.

1157

1158 M. R. SARKISIAN ET AL.

While many studies have demonstrated the long-term age-related adverse effects of a single prolonged seizure, it is less clear whether there are similar age-related dif- ferences in the cognitive and pathologic sequelae after recurrent or repetitive seizures. Since many children and adults experience recurrent seizures despite treatment with antiepileptic drugs (AEDs), this is a question of criti- cal clinical importance. To determine whether there are age-related differences in cognitive outcome following repetitive seizures, we evaluated visuospatial learning and memory in immature and mature rats following mul- tiple injections of KA. We examined the hypothesis that multiple KA seizures in the immature brain would lead to pathological changes and behavioral deficits similar to those seen in adult rats after a single KA-induced episode of status epilepticus, but not after a single KA seizure in immature rats. However, we found that the immature brain appears to be protected from the deleterious effects of KA even after 4 injections at convulsive doses. These findings suggest an inherent resistance of the immature brain to KA seizure-induced brain damage.

METHODS Experimental design

Litters of Sprague-Dawley male rats were purchased from Charles River Laboratories, Cambridge, MA, U.S.A. At postnatal age (P, in days) P15 (immature) or P55 (mature), rats underwent water maze training con- sisting of 1 day of habituation followed by 4 days of place training to learn to find the platform and escape onto it. Following completion of water maze training, rats were assigned to KA or control groups (P20: 8 con- trols, 16 KA-treated; P60: 7 controls, 12 KA-treated), with groups balanced as to water maze performance. Rats then received serial injections of KA at 2-day in- tervals. The immature group received intraperitoneal (ip) injections of KA (10 mgkg/day) or saline on P20, P22, P24, and P26, while the mature group received KA (14 mg/kg/day) or saline on P60, P62, P64, and P66. Age- appropriate KA doses were chosen to cause severe status epilepticus with a mortality rate of less than about 25% (9). Controls received equal volume injections of saline. Immediately following an injection, rats were placed into individual plastic cages and their behavior was scored every 15 min for 2 h, noting the latency-to-onset of the various KA seizure stages characteristic of each age (see Results). Following the observation period, animals were returned to their home cages for 2 days, until the next KA injection. To assess the long-term behavioral effects of repetitive KA seizures, water maze testing was repeated in the immature group beginning at P60, and in the ma- ture group beginning at P100. Water maze

A circular pool made from a galvanized stock water tank (1 17 cm diameter x 50 cm high) was filled to a

depth of 25 cm with water at a temperature of 26 * 1°C. The pool was illuminated by overhead fluorescent lights and kept in a permanent location throughout the study. Four points on the rim of the pool were designated as north, south, east and west. On day one, the rats were placed in the pool for 60 seconds of habituation to the apparatus with no platform present (“free swim”). On days 2 through 5, rats were trained for 24 trials (6 trials per day) to locate and escape onto a plexiglass platform (8 x 8 cm) placed 1.5 cm under the water surface. The platform was kept in the same location throughout the study, for both pre- and post-KA training. The water was made opaque by the addition of evaporated milk. At the start of each trial, the rat was held facing the perimeter and dropped into the pool to ensure immersion. Entry from the N, S, E or W points was varied in a quasi- random order. Time to escape onto the platform and swim route were recorded manually. The study design tests the ability of an animal to learn and remember the spatial location of the escape platform using only visual cues, and assesses hippocampal integrity (10,ll). EEG

To monitor EEG changes following serial injections of KA, another group of rats at P16 (n = 5) and P56 (n = 6) had electrodes implanted into the dorsal hippocampus. Animals were anesthetized with chloral hydrate (500 mg/ kg, ip) and electrodes were implanted stereotaxically into the dorsal hippocampus, using the following coordinates (in cm): AP = -0.04, ML = 0.45, and DV = -0.4 for P16 rats (reference to interaural line), and AP = -0.23, ML = 0.10, and DV = -0.42 for P56 animals (12). Following a 4-day recovery period, the animals received serial injections of KA, using the same dosages and schedule as described above. EEGs were recorded for 2 min prior to KA and for 2 h after each KA injection. These rats were not subjected to water maze testing.

Histology Upon the completion of the second water maze series

(i.e., post-KA injections), rats were killed with a lethal dose of sodium pentobarbital (50 mgkg), and decapi- tated. Brains were rapidly removed and frozen by im- mersion in liquid nitrogen. The frozen brains were sec- tioned on a cryostat (-17°C) at 20 pm in the coronal plane. Every fifth section containing the hippocampus was mounted onto slides and stained with cresyl violet, analyzed for major cell loss, and photographed. Qualita- tive comparisons of hippocampal neuronal loss were made between controls and KA-treated rats in each age group.

Statistics Two-way analysis of variance (ANOVA) with re-

peated measures was used to compare water maze escape latencies over the 24 trials of place learning. One-way ANOVA was used to analyze latencies to various seizure

Epilepsia, Vol. 38, No. 11, 1997

REPEATED KA DURING DEVELOPMENT 1159

stages. Student's t test was used to compare body weight changes. Chi square statistics were used for analysis of frequency data. Significance levels was defined as p < 0.05 for all comparisons.

RESULTS

Ictal and electrographic characteristics of multiple KA seizures

Behavioral changes in response to the first of a series of KA injections on P20 were similar to those described previously (3,9,13). Rats became immobile, with staring and ataxia, and later developed wet-dog shakes, alternate limb scratching and eventually forelimb clonus (synchro- nous or asynchronous). In the older rats, KA administra- tion on P60 caused the classic sequence of KA seizures, beginning with immobility and progressing through wet- dog shakes, salivatiodfacial clonus, unilateral and then bilateral forelimb clonus, and finally, rearing and falling (13,14). Forelimb clonus is a hallmark of progression of seizure activity beyond limbic structures to involve mo- tor cortex, while rearing and falling suggest that the sei- zure has generalized.

However, the response to serial injections of KA was considerably different in the 2 age groups. In rats which received injections staring at P20, there was a progres- sive decrease in seizure severity and an increase in the latency-to-seizure onset, with each subsequent injection. As seen in Fig. lA, 68% of immature rats developed a seizure with forelimb clonus after the first injection of KA on P20. Thereafter, following subsequent injections, progressively fewer rats developed forelimb clonus. The second and third KA administrations (P22, P24) caused staring and wet-dog shakes in most rats, but the forelimb clonus stage was mild and brief and occurred in only 21% and 26% of rats, respectively (Fig. 1A). After 90 min, most rats were noted to be asleep. On the fourth day (P26), none of the rats showed any seizure sign: instead, they all fell asleep. After several hours, once awake, the behavior of KA-injected immature rats appeared normal. None of the immature rats began seizing more than 120 min after KA, and none died after any of the KA injec- tions.

Among rats which received serial KA injections as adults, beginning at P60,72% developed forelimb clonus on day 1 (P60), 58% on day 2 (P62) and 82% on day 3 (P64) (Fig. 1B). On the fourth day of KA injection (P66), only 30% of rats reached forelimb clonus. In the adult group, there was a mortality rate of 33% following the first injection, 12.5% following the second injection, 14% following the third injection, and 0% following the fourth injection. In adults, there was a variable pattern of seizure latency and severity with serial injections.

Frequency analysis was performed on rats developing a seizure in response to KA after each of the four injec- tions, comparing immature and adult rats. In the imma-

A Immature

loo 75 1 Ye 50

25

0

KA#1 KA#2 KA#3 KA#4 P20 P22 P24 P26

B Adult loo 1

KA#1 KA#2 KA#3 KA#4 P60 P62 P64 P66

FIG. 1. Percentage of rats reaching the seizure stage of fore- limb clonus (FLC) following each KA administration in immature (A) and adult (B) rats. Day of KA injection and age at each KA administration are indicated on the abscissa. In A, no rat reached FLC after the 4th injection.

A

Latency (fin)

B

Latency

30 20 'I 10 0

30 20

Immature

T

KA#l KA#2 KA#3 KA#4 P20 P22 P24 p26

Adult T

KA#1 KA#2 KA#3 KA#4 P60 P62 P64 P66

FIG. 2. Latency to forelimb clonus following serial injections of KA beginning at P20 (A) or P60 (8). Among immature rats, there was a progressive increase in FLC latency. Latency to FLC after the 3rd KA injection was significantly longer than latencies after the first or second KA administration; no FLC occurred after the fourth KA injection. In adult rats (B), there were no significant differences in FLC latency at any day of KA injection. Error bars indicate S.E.M.

Epilepsia, Vol. 38, No. 11, 1997

1160 M. R. SARKISIAN ET AL.

ture group, significantly fewer rats developed seizures in response to KA with subsequent doses (x2(3, = 22.8, p < 0.0001). In contrast, there was no significant difference in the number of adult rats developing seizures in re- sponse to subsequent injections of KA (x20) = 4.8, p = 0.25). Furthermore, among rats which did develop sei- zures in response to serial injections of KA (i.e., exclud- ing those which did not seize), there was a significant in- crease in latency to forelimb clonus with subsequent in- jections in immature but not adult rats (one-way ANOVA; immature: F1,,* = 4.28, p = 0.03; adult: F,,,, = 2.82, p = 0.06) (Fig. 2). Therefore, fewer immature rats de- veloped seizures with each successive KA dose and the latency to forelimb clonus progressively increased.

EEG changes following multiple KA seizures Ictal discharges, defined as rhythmic electrographic

waveforms with sharp morphology, occurred on all 4

days of KA injection in both P20 and P60 groups. How- ever, in the immature rats, the ictal discharges on the 3rd and 4th KA days were sporadic and poorly sustained compared to the older rats, in which ictal discharges were virtually continuous on each day of injection, even in those adult rats which had minimal or no overt clinical seizure activity (Fig. 3).

Water maze performance before and after multiple KA seizures

Before KA, at each age, rats performed similarly in learning the position of the platform in the water maze. Therefore, saline vs. KA groups were balanced with re- gard to water maze performance. There were no signifi- cant differences in the learning position of the platform between animals receiving serial injections of saline or KA beginning on P20 (Fig. 4A). It is noteworthy that even at such a young age, rats eventually learn the plat-

A

5 Sec.

F"22

B

P64

FIG. 3. Representative EEG traces fol- lowing each of the four serial injections of KA in immature (A) and adult (B) rats. In the immature rats, high voltage irregular spikes are seen after the first KA injection (P20), and less continuous, less frequent discharges occur on subsequent days. In adults, continuous ictal discharges are seen after each KA injection, even when clinical ictal changes were not prominent (e.g., P66).

Epilepsia. Vol. 38, No. 11, 1997

REPEATED KA DURING DEVELOPMENT

500 h y 450 -

2 400 - of 2 350 -

??

I 1 6 1

FIG. 4. Latencies to escape onto the water maze platform in immature (A) and adult (B) rats. Indicated on the abscissa are the ages of water maze place training (P16-19 in A; P56-59 in B), the ages of KA or sa- line administration (arrow) and the ages at the time of repeat water maze testing (P60-63 in A; PIOO-103 in B). Error bars indicate +SEM.

A Long-term Effect of Repeated KA - Immature Rats

I- 4 300 ] a

W

I- 150 1 2 200

100

50 &- I I I I 7

t PI6 PI7 PI8 P I9

B

250

r ; 200

4 a 150

P W

100 F

50

Repeated Saline

*-a Repeated KA

T

P20,22,24,+26 - Repeated KA or Saline Injections

Adult Rats

&-A Repeated Saline

Repeated KA

i \

\

0 1 I I I I I I I I I I

PlOO PI01 PI02 PI03 1\ P56 P57 P58 P59

P60,62,64,+66 - Repeated KA or Saline Injections

form position, although the latencies are much longer than for older rats (Fig. 4A, left panel). When these im- mature rats were later retested in the water maze as adults (beginning at P60), the saline and KA-treated groups performed equivalently well in learning the spatial loca- tion of the platform (Fig. 4A, right panel), with latencies strikingly similar to the pre-KA adult group, which was tested from P56-P59 (compare with Fig. 4B, left panel).

Following serial KA, the age groups differed markedly in their water maze performance. In rats receiving KA

beginning at P60, there was a significant impairment in the latency to find the platform compared to controls (F1,lB = 11.3; p < 0.001) (Fig. 4B, right panel), whereas no differences were seen in the immature group (Fig. 4A, right panel). Importantly, saline-treated adult rats re- tained memory of platform position from P59-P100 (de- spite repeated saline injections) whereas repeated KA treatments caused loss of spatial memory during this in- terval (compare filled hexagons at P59 and PlOO in Fig. 4B).

Epilepsia, Vol. 38, No. 11, 1557

1162 M . R. SARKISIAN ET AL.

Effects of multiple KA injections on body weight Immature rats treated with multiple injections of KA

or saline gained weight similarly over the injection pe- riod, with weights in each group increasing by about 55% from P20 to P26. However, the group of adult rats treated with multiple KA injections from P60 to P66 lost 11% of their body weight over the injection period, while the saline-treated group of adult rats gained 25% (p = 0.0036, t test). Therefore, multiple KA treatments in im- mature rats do not affect feeding behavior as assessed by a normal pattern of weight gain, as opposed to the situ- ation in adult rats, where serial KA led to progressive weight decline over the period of KA injections.

Histological analysis Histological examination revealed marked cell loss in

the CA1, CA3, CA4, and hilar regions of the hippocam-

FIG. 5. Cresyl violet-stained hip- pocampal sections from rats treated with repeated saline from P20-26 (A), repeated KA from P20-26 (B), repeated saline from P60-66 (C) , and repeated KA from P60-66 (D). Note normal hippocampal structure in A, B, and C, with no obvious neuronal damage observable. In D, how- ever, there is nearly total cell death in all subfields except CA2. Scale bar (400 vm) applies to all panels.

pus in all 6 surviving rats (100%) treated with repeated KA from P60-P66 (Fig. 5D). In addition, in 5 of these 6 rats, there was apparent dispersion of granule cells in the dentate gyms. No major lesions were detected in rats treated with repeated KA from P20-26 or in saline- treated rats at either age (Fig. 5A,B,C). In rats which had had EEG electrodes implanted, the presence of the elec- trode tract was verified histologically in the dorsal hip- pocampus.

DISCUSSION

In this study, we examined the hypothesis that mul- tiple KA seizures in the immature brain would lead to behavioral deficits and pathological changes similar to those seen in the adult brain following a single prolonged KA seizure. However. we found that the immature brain

Epilepsia. Vol. 38, No. 11, 1997

REPEATED KA DURING DEVELOPMENT

FIG. 5. Continued.

1 I63

is protected from the detrimental effects of KA even after 4 injections of the convulsant. These results suggest an inherent resistance of the immature brain to KA seizure- induced brain damage, findings that may have important clinical implications.

We found that the immature brain is protected against 3 major adverse sequelae of multiple KA exposures: be- havioral and electrographic seizure manifestations, im- pairment of visuospatial learning and memory, and his- tologic damage to hippocampal pyramidal and hilar neu- rons. Immature rats exhibit progressively fewer severe seizures following each of 4 intraperitoneal injections of KA, spaced 2 days apart. This result was contrary to our expectation that serial KA administration would cause progressively more severe status epilepticus with each injection. Immature rats learned to find the platform in

the water maze even at young ages (P16-P19), and when tested as adults they retained memory of the platform location even after 4 exposures to convulsant doses of KA. The progressive increase in body weights during the experimental period in both saline and KA-treated im- mature rats attests to the lack of effect of serial KA injections on growth. Furthermore, there were no histo- logic lesions in the hippocampus after 4 KA exposures.

These results are in stark contrast to our findings in adult rats: In most rats, 4 serial KA injections caused a severe seizure each day, as judged by both behavioral and electrographic criteria. KA-treated adults decreased their food intake and lost weight progressively over the experimental period, although their intake increased again after the acute effects of KA wore off. Adults treated with 4 KA injections had major cognitive impair-

1164 M. R. SARKISIAN ET AL.

ment, a finding judged by water maze performance. They also exhibited severe neuronal damage in the hippocam- pus involving subfields CA1, CA3 and CA4. In addition, there was apparent widening of the dentate granule cell layer, possibly suggesting dispersion of these cells. How- ever, we did not perform cell-counting, and an alterna- tive hypothesis is that the number of granule cells actu- ally increased in KA-treated rats, similar to the increase in dentate granule neurons seen after prolonged seizures induced by pilocarpine or perforant path stimulation in adult rats (15). It is our impression that the histological damage seen after 4 KA injections is more severe than those seen after a single episode of KA-induced status epilepticus, but serial sacrifice of animals was not per- formed in this study.

The differences between immature and adult rats in their sensitivity to the effects of multiple KA injections cannot be explained by the differences in the KA dose used at the 2 ages. At both ages, the dose was selected to cause severe status epilepticus based on previous studies (3,9,13). All these studies confirm that a lower dose of KA is sufficient to induce status epilepticus in immature rats compared to adults, which require a higher dose. This dose-sensitivity difference might be explained by either an immature blood brain barrier, or increased sen- sitivity to KA, or both. Use of the higher dose (14 mgkg) in immature rats would certainly have caused an unac- ceptably high mortality rate. That the doses chosen here produced equivalent seizures at the 2 ages is suggested by Fig. 1, where roughly 70% of both immature and adult rats developed forelimb clonus after the initial KA injection .

We have shown previously that serial daily adminis- tration of KA to rats beginning at P25, at doses ranging from 0.25-7.0 mgkg, did not result in “pharmacologic kindling;’ ’ that is, there was no intensification of seizure severity with serial doses (16). Rather, subsequent ad- ministrations of KA resulted in less severe seizures, simi- lar to the “tolerance” observed in the present study. These findings suggest that KA acts differently than some other convulsants, which may produce a “kindling effect” (17,18).

It is unclear why such a remarkable age difference is seen following serial KA administration. It has been pos- tulated that the immature brain is protected against the detrimental effects of KA because of an incomplete glu- taminergic (mossy fiber) input pathway from dentate granule cells to CA3 pyramidal neurons and cells of the dentate hilus (2,5,19). In the rat, this pathway is suppos- edly mature by about P21; in our experiments, KA was administered between P20 and P26, ages when at least some KA-induced damage is expected. KA receptors are present and functional at this age (20), and since pro- longed KA-induced seizures do occur in rats at this age, we can assume that the KA receptors are activated by the

KA and that glutamate is present and released. Since KA acts in part by releasing glutamate from presynaptic stores (21), it is possible that excessive KA-released glu- tamate may be playing a protective role against seizure- induced damage. Results from our laboratory indicate that direct injections of glutamate into the hippocampus of P10 rats failed to produce cell loss, while injection of the same glutamate dose into hippocampi of P30 or P60 rats caused extensive focal neuronal loss; P20 rats were intermediate in the degree of glutamate-induced necrosis (22). Moreover, glutamate has been shown to promote the survival of dentate granule cells during development (23), while glutamate and its agonists enhance the sur- vival of neurons at low concentrations (24-27). These effects of glutamate might also be linked to the lack of neuronal death following repeated KA administration, with increased glutamate levels being neuroprotective rather than excitotoxic at some levels or at certain de- velopmental stages.

Alternatively, the protective mechanism might be re- lated to altered KA receptor regulation. That is, there may be structural or functional downregulation of KA receptors with subsequent KA administrations, to such a degree that fewer receptors are available to mediate ei- ther a seizure or excitotoxicity. Such a downregulation of neurotransmitter receptors is seen in other pathways (28- 30), and might help to explain the decrease in seizure intensity we observed with repeated KA injections in developing animals. Downregulation of receptors in other neurotransmitter systems (e.g., cholinergic) has been shown to produce “tolerance” (31,32). A similar tolerance mechanism may be operative in our study. We have preliminary evidence that repeated KA administra- tion to immature rats but not adults causes downregula- tion of KA receptor number (33, and our unpublished observations). However, it is unknown why such toler- ance would occur only in the immature brain.

Another possibility is that in the immature brain, KA seizures may be associated with an endogenous protec- tive agent such as neurotrophins. Several reports have suggested that alterations in neurotrophins (e.g., nerve growth factor, NGF; basic fibroblast growth factor, bFGF; brain-derived growth factor, BDNF) occur fol- lowing seizures in adult animals (34-36), a protective mechanism which might play an even more prominent role in the immature brain (37,38). The developing brain has higher levels of neurotrophins than does the adult brain, and high concentrations of NGF, BDNF, bFGF and neurotrophins 3 and 4 (NT3, NT4) are needed for the development and support of various neuronal popula- tions during development (39-43). In the adult, noninju- rious neuronal stimulation and kindling stimulations have been reported to increase the levels of these neu- rotrophins (34-36). Stimulation of excitatory amino acid (glutamate) receptors leads to an increase in BDNF

Epilepsiu, Vol. 38, No. 11, 1997

REPEATED KA DURING DEVELOPMENT 1165

(44,45), which in turn protects neurons from excitotoxic death. It is possible that repeated KA administrations produce a neuroprotective effect by permitting a pro- longed increase of neurotrophin levels, and that this may be more beneficial to the developing brain. Our prelimi- nary results suggest that blocking BDNF synthesis by chronic infusion of the antisense deoxyoligonucleotide to BDNF is associated with KA-induced CAI and CA3 damage in P20 rat hippocampus (an age at which KA does not ordinarily produce such hippocampal damage)

Similar neuroprotective mechanisms following sei- zures have been proposed previously. Rats kindled from the hippocampus had less severe behavioral and electri- cal status epilepticus and less neuronal damage in the piriform cortex and hippocampus in response to subse- quent KA than did naive (non-kindled) rats (47). In an- other study, KA-induced status epilepticus terminated with pentobarbital (PTB), then restarted with another KA dose 16 hours later, showed protection of CA3 neurons (48).

We have demonstrated previously that immature rats (P5-30) receiving a single dose of KA sufficient to pro- duce status epilepticus have no impairment in water maze position learning when tested later as adults (6). Taken together with the present findings, we can con- clude that the immature brain is less vulnerable to the long-term cognitive effects of either a single or recurrent seizures than the mature brain. These findings may have relevance to epilepsy in patients. In children, multiple recurrent seizures are common, and it is often assumed that repetitive seizures lead to brain damage in the child (49). It must be acknowledged that while our results and those of others suggest some degree of protection from seizures early in development, under certain conditions the immature brain is not completely immune to seizure- induced damage. Both prolonged (50,5 1) and recurrent seizures (52) can lead to deficits during development, and kindling in immature animals can result in a lasting increase in seizure susceptibility (53,54). Many factors can contribute to seizure-induced damage, including the method of seizure induction, age of the animal, seizure type, duration, and frequency. Each of these variables must be considered when assessing the role of seizures in neurologic sequelae.

Our findings should not, of course, be interpreted as suggesting that seizures may be beneficial to the imma- ture brain. Instead, our results raise the possibility that there may be age-specific protective mechanisms in the developing brain to ameliorate seizure-induced damage. Further elucidation of the mechanisms leading to age- related protection from or susceptibility to seizure- induced damage could lead to the development of novel age-specific antiepileptic therapies.

(46).

Acknowledgment: The authors are grateful for the support of this study by NINDS grant (NS27984) to GLH, and a Basil O’Connor Starter Scholar Award from the March of Dimes Birth Defects Foundation to CES.

REFERENCES

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

Nadler JV. Kainic acid as a tool for the study of temporal lobe epilepsy. Life Sci 1981;29:2031-42. Ben-Ari Y . Limbic seizure and brain damage produced by kainic acid Mechanisms and relevance to human temporal lobe epilepsy. Neuroscience 1985; 14:375-403. Albala BJ, Moshe SL, Okada R. Kainic acid-induced seizures: A developmental study. Dev Bruin Res 1984;13: 13943. Nitecka L, Tremblay E, Charton G, Bouillot JP, Berger ML, Ben- Ari Y. Maturation of kainic acid seizure-brain damage syndrome in the rat: 11. Histopathological sequelae. Neuroscience 1984; 13: 1073-94. Sperber EF, Haas KZ, Stanton PK, Mosht SL. Resistance of the immature hippocampus to seizure-induced synaptic reorganization. Dev Bruin Res 1991;60:88-93. Stafstrom CE, Holmes GL, Chronopoulos A, Thurber S, Thomp- son JL. Age-dependent cognitive and behavioral deficits following kainic acid-induced seizures. Epilepsia 1993;34:420-32. Thurber S, Chronopoulos A, Stafstrom CE, Holmes GL. Behav- ioral effects of continuous hippocampal stimulation in the devel- oping rat. Dev Bruin Res 1992;68:3540. Liu Z , Gatt A, Werner SJ, Mikati MA, Holmes GL. Long-term behavioral deficits following pilocarpine seizures in immature rats. Epilepsy Res 1994;19:191-204. Stafstrom CE, Thompson JL, Holmes GL. Kainic acid seizures in the developing brain: Status epilepticus and spontaneous recurrent seizures. Dev Brain Res 1992;65:227-36. Morris RGM, Garrud P, Rawlins JNP, O’Keefe J. Place navigation is impaired in rats with hippocampal lesions. Nature 1982;297: 681-3. Sutherland RJ, Whishaw IQ, Kolb B. A behavioral analysis of spatial localization following electrolytic, kainate or colchicine- induced damage to the hippocampus. Behav Brain Res 1983;7:

Shenvood NM, Timiras PS. A Stereotuxic Atlas of the Developing Rut Bruin. Berkeley, CA: University of California Press, 1970. Tremblay E, Nitecka L, Berger ML, Ben-Ari Y. Maturation of kainic acid seizure brain damage syndrome in the rat: I. Clinical, electrographic and metabolic observations. Neuroscience 1984;13: 1051-72. Sperk G. Kainic acid seizures in the rat. Progr Neurobiol 1994; 42: 1-32. Parent JM, Yu TW, Leibowitz RT, Geschwind DH, Sloviter RS, Lowenstein DH. Dentate granule cell neurogenesis is increased by seizures and contributes to aberrant network reorganization in the adult rat hippocampus. J Neurosci 1997;17:3727-38. Holmes GL, Thompson JL. Effects of serial administration of kainic acid on the developing brain. Neuropharmucology 1988;27: 209-12. Vosu H, Wise RA. Cholinergic seizure kindling in the rat: Com- parison of caudate, amygdala and hippocampus. Behuv Biol 1975; 13:491-4. Diehl RG, Smialowski A, Gotwo T. Development and persistence of kindled seizures after repeated injections of pentylenetetrazole in rats and guinea pigs. Epilepsiu 1984;25:50&10. Ribak CE, Navetta MS. An immature mossy fiber innervation of hilar neurons may explain their resistance to kainate-induced cell death in 15-day-old rats. Dev Bruin Res 1994;79:47-62. Miller LP, Johnson AE, Gelhard RE, Insel TR. The ontogeny of excitatory amino acid receptors in rat forebrain: 11. Kainic acid receptors. Neuroscience 1990;35:45-5 1. Chittajallu R, Vignes M, Dev KK, Barnes JM, Collingridge GL, Henley JM. Regulation of glutamate release by presynaptic kainate receptors in the hippocampus. Nature 1996;379:78-81. Liu 2, Stafstrom CE, Sarkisian M, et al. Age-dependent effects of

133-53.

Epilepsia, Vol. 38, No. 11, 1997

1166 M. R. SARKISIAN ET AL.

glutamate toxicity in the hippocampus. Dev Bruin Res 1996;97: 17 8-84.

23. Gould E, Cameron HA, McEwen BS. Blockade of NMDA recep- tors increases cell death and birth in developing rat dentate gyms. J Comp Neurol 1994;340:55145.

24. Balazs R, Jorgensen OS, Hack N. N-methyl-D-aspartate promotes the survival of cerebellar granule cells in culture. Neuroscience

39. Kaisho Y, Shintani A, Ono Y, Kato K, Igarashi K. Regional ex- pression of the nerve growth factor gene family in rat brain during development. Biochem Biophys Res Commun 1991 ; 174:379-85.

40. Nonner D, Barrett JN. Changes in the response of cultured septa1 cholinergic neurons to nerve growth factor exposure and depriva- tion during the first postnatal month. Dev Bruin Res 1994;79:219- ?P

25.

26.

27.

28.

29.

30.

31

32.

33.

34.

35.

36.

37.

38.

- 1988;27:437-51. Yan GM, Ni B, Weller M, Wood KA, Paul SM. Depolarization or glutamate receptor activation blocks apoptotic cell death of cul- tured cerebellar granule neurons. Bruin Res 1994;656:43-52. Bambrick LL, Yarowski PL, Krueger BK. Glutamate as a hippo- campal neuronal survival factor: An inherited defect in the trisomy 16 mouse. Proc Nut1 Acud Sci USA 1995;92:9692-6. Suzuki F, Junier M-P, Guilhem D, Sorenson J-C, Onteniente B. Morphogenetic effect of kainate on adult hippocampal neurons associated with a prolonged expression of brain-derived neuro- trophic factor. Neuroscience 1995;64:665-74. Kang I, Miller LG. Decreased GABA, receptor subunit mRNA concentrations following chronic lorazepam administration. Br J Phurmacol 1991;103:1285-7. Temple A. Visualization of mu opiate receptor downregulation following morphine treatment in neonatal rat brain. Bruin Res

Haas KZ, Sperber EF, MoshC SL, Stanton PK. Kainic acid-induced seizures enhance dentate gyrus inhibition by downregulation of GABA, receptors. J Neurosci 1996; 16:425O-60. Bushnell PJ, Padilla SS, Ward T, Pope CN, Olszyk VB. Behavioral and neurochemical changes in rats dosed repeatedly with diisopro- pylfluorophosphate. J Pharmucol Exp Ther 1991;256:741-50. Clement JG. Effect of a single dose of acetylcholinesterase inhibi- tor on oxotremorine- and nicotine-induced hypothermia in mice. Pharmacol Biochem Behuv 1991;39:929-34. Tandon P, Sarkisian MR, Stafstrom CE, Yang Y, Holmes GL. Repeated exposure to kainic acid in developing rats does not cause repeated seizures, neuronal loss or behavioral deficits later in life. Epilepsia 1996;37[Suppl. 5]:66. Ballarin M, Emfors P, Lindefors N, Persson H. Hippocampal dam- age and kainic acid injection induce a rapid increase in mRNA for BDNF and NGF in the rat brain. Exp Neurol 1991;114:35-43. Riva MA, Gale K, Mocchetti I. Basic fibroblast growth factor mRNA increases in specific brain regions following convulsive seizures. Molec Brain Res 1992;15:311-8. Follesa P, Gale K, Mocchetti I. Regional and temporal pattern of expression of nerve growth factor and basic fibroblast growth fac- tor mRNA in rat brain following electroconvulsive shock. Exp Neurol 1994;127:37-44. Dugich-Djordjevic MM, Tocco G, Willoughby DA et al. BDNF mRNA expression in the developing rat brain following kainic acid-induced seizure activity. Neuron 1992;8: 1127-38. Larmet Y, Reibel S, Carnahan J, Nawa H, Marescaux C, Depaulis A. Protective effects of brain-derived neurotrophic factor on the development of hippocampal kindling in the rat. Neuroreport 1995;6: 1937-41.

1991;64: 19-26.

1 v .

41. Zhou J, Bradford HF, Stern G. The stimulatory effect of brain- derived neurotrophic factor on dopaminergic phenotype expression of embryonic rat cortical neurons in vitro. Dev Bruin Res 1994; 81:3 18-24.

42. Thoenen H. Neurotrophins and neuronal plasticity. Science 1995; 270:593-8.

43. Zhou X-F, Rush RA. Sympathetic neurons in neonatal rats require endogenous neurotrophin-3 for survival. J Neurosci 1995;15: 6521-30.

44. Zafra F, Hengerer B, Leibrock J, Thoenen H, Lindholm D. Activ- ity-dependent regulation of BDNF and NGF mRNAs in the rat hippocampus is mediated by nonNMDA glutamate receptors. EMBO J 1990;9:3545-50.

45. Wetmore C, Olson L, Bean AJ. Regulation of brain derived neu- rotrophic factor (BDNF) expression and release from hippocampal neurons is mediated by non-NMDA type glutamate receptors. J Neurosci 1994;14: 1688-700.

46. Tandon P, Sarkisian M, Yang Y, et al. The neuroprotective role of BDNF in seizure induced neuronal loss during development. Neu- rology 1997;48:A15 1.

47. Kelly ME, McIntyre DC. Hippocampal kindling protects several structures from the neuronal damage resulting from kainic acid- induced status epilepticus. Bruin Res 1994;634:245-56.

48. Najm I, Schreiber S S , Bruce A, Tocco G, Baudry M. A short episode of seizure activity protects CA3 neurons from prolonged seizure activity-induced death. Soc Neurosci Abstr 1992;18: 1146.

49. Holmes GL. Do seizures cause brain damage? Epilepsia 1991;32[Suppl. 5]:S14-28.

50. Frank JE, Schwartzkroin PA. Immature rabbit hippocampus is damaged by systemic but not intraventricular kainic acid. Dev Bruin Res 1984;13:219-27.

51. Babb TL, Pereira-Leite J, Mathem GW, Pretorius JK. Kainic acid- induced hippocampal seizures in rats: Comparisons of acute and chronic seizures using intrahippocampal versus systemic injec- tions. Itul J Neurol Sci 1995;16:3944.

52. Neil1 JC, Liu Z, Sarkisian M, et al. Recurrent seizures in immature rats: Effect on auditory and visual discrimination. Dev Bruin Res 1996;95:283-92.

53. Mosh6 SL, Albala BJ. Kindling in developing rats: Persistence of seizures into adulthood. Dev Brain Res 1982;4:67-71.

54. MoshC SL, Albala BJ. Maturational changes in postictal refracto- riness and seizure susceptibility in developing rats. Ann Neurol 1983;13:552-7.

Epilepsia, Vol. 38, No. 11. 1997